An initial assessment of discriminant surface complexity for power law features

Abstract
The detection of man-made objects in natural terrain is important in both the targeting and terminal homing phase of modern warfare. The presence of man-made objects in gray-scale images has been successfully detected using a new class of density estimation neural networks to analyze power law signatures. The complex nature of the discriminant surface relating these features has been elucidated using these adaptive mixture networks.

This publication has 11 references indexed in Scilit: